First-Line Toxicological Screening with Fully Automated Extraction.

In clinical toxicology, laboratories need screening methods allowing unambiguous identification of the compounds in a short turnaround time to either confirm or exclude the hypothesis of drug overdose or poisoning with a toxicant. We developed a fully automated screening procedure designed to identify and quantify in a single run 245 compounds of interest in clinical toxicology. Sample extraction was carried out by a programmable liquid handler directly coupled to a liquid chromatography-tandem mass spectrometry (LC-MS-MS) system. Data acquisition was performed in the positive and negative ionization modes with up to 15 multiple reaction monitoring (MRM) transitions per compound, each with optimized collision energy to enable both qualitative library searching and quantitation. The method was validated according to the ISO 15189 requirements and was applied to real patient samples (n = 127). The 15 MRM transitions per compound provided higher confidence for the identification of all the compounds. The quantitative method was fully validated with satisfactory intra- and inter-assay imprecision and inaccuracy with CV% lower than 20%. For only nine molecules, imprecision and inaccuracy were relatively high but never exceeded 31.7%. Comparison with dedicated quantitative methods using conventional MRM monitoring performed using 127 patient samples (n = 175 pairs of measured concentrations) showed excellent correlation (R2 = 0.96). A robustness study showed that calibration curves prepared for up to 1 month yielded uncertainty < 20%. Retention times ranged from 0.89 min for metformin to 9.72 min for difenacoum. The automated sample preparation required 8 min and was followed by 10 min chromatographic separation. This first-line screening procedure yields high confidence in compound detection and should be useful in core labs facing clinical toxicology situations where rapid and reliable results are needed.

Automatic quantification of uracil and dihydrouracil in plasma.

Fluoropyrimidines-based chemotherapies are the backbone in the treatment of many cancers. However, the use of 5-fluorouracil and its oral pre-prodrug, capecitabine, is associated with an important risk of toxicity. This toxicity is mainly due to a deficiency of dihydropyrimidine dehydrogenase (DPD). This deficiency may be detected by using a phenotypic approach that consists in the measurement of uracilemia or the calculation of dihydrouracil (UH2)/uracil (U) ratio. For uracilemia, a threshold value of 16 ng/ml has been proposed for partial deficiency, while a value of 150 ng/ml has been proposed for complete deficiency. We have developed a rapid, accurate and fully-automated procedure for the quantification of U and UH2 in plasma. Sample extraction was carried out by a programmable liquid handler directly coupled to a liquid chromatography - tandem mass spectrometry (LC-MS/MS) system. The method was validated according to the EMA guidelines and ISO 15189 requirements and was applied to real patient samples (n = 64). The limit of quantification was 5 and 10 ng/ml for U and UH2 respectively. Imprecision and inaccuracy were less than 15% for inter and intra-assay tests. Comparison with dedicated routine method showed excellent correlation. An automated procedure perfectly fulfills the need of low inaccuracy and CVs at the threshold values (less than 5% at 16 ng/ml) and is highly suitable for the characterization of DPD deficiency. Automatization should guaranty reliable and robust performances by minimizing the sources of variation such as volume inaccuracies, filtration or manual extraction related errors.

Fully automated sample preparation procedure to measure drugs of abuse in plasma by liquid chromatography tandem mass spectrometry.

For the analysis of drugs and pharmaceutical compounds in biological matrices, extraction procedures are typically used for LC-MS/MS analysis often requiring manual steps in sample preparation. In this study, we report a fully automated extraction method carried out by a programable liquid handler directly coupled to an LC-MS/MS system for the determination of 42 components (illicit drugs and/or metabolites) (plus 20 deuterated internal standards). The acquisition was performed in positive ionization mode with up to 15 MRM transitions per compound, each with optimized collision energy (MRM spectrum mode) to enable qualitative library searching in addition to quantitation. After placing the sample tube into the system, no further intervention was necessary: automated preparation used 50 µL of blood or plasma with 3 µL of extracted sample injected for analysis. The method was validated according to the requirements of ISO 15189. The limit of detection and quantification was 1-5 ng/mL depending on the compound. Stability experiments found that historic calibration curve data files could accurately quantify for up to 1 month with less than 20% uncertainty. Comparison to a QuEChERS method was made using patient samples providing a regression correlation R2 = 0.98 between the two methods. This approach was successfully designed to support parallel sample preparation and analysis therefore significantly increasing sample throughput and reduced cycle times. Graphical abstract ᅟ.